Abstract: Effectiveness of Artificial Neural Networks (ANN)
and Support Vector Machines (SVM) classifiers for fault diagnosis of
rolling element bearings are presented in this paper. The
characteristic features of vibration signals of rotating driveline that
was run in its normal condition and with faults introduced were used
as input to ANN and SVM classifiers. Simple statistical features such
as standard deviation, skewness, kurtosis etc. of the time-domain
vibration signal segments along with peaks of the signal and peak of
power spectral density (PSD) are used as features to input the ANN
and SVM classifier. The effect of preprocessing of the vibration
signal by Discreet Wavelet Transform (DWT) prior to feature
extraction is also studied. It is shown from the experimental results
that the performance of SVM classifier in identification of bearing
condition is better then ANN and pre-processing of vibration signal
by DWT enhances the effectiveness of both ANN and SVM classifier
Abstract: As computer network technology becomes
increasingly complex, it becomes necessary to place greater
requirements on the validity of developing standards and the
resulting technology. Communication networks are based on large
amounts of protocols. The validity of these protocols have to be
proved either individually or in an integral fashion. One strategy for
achieving this is to apply the growing field of formal methods.
Formal methods research defines systems in high order logic so that
automated reasoning can be applied for verification. In this research
we represent and implement a formerly announced multicast protocol
in Prolog language so that certain properties of the protocol can be
verified. It is shown that by using this approach some minor faults in
the protocol were found and repaired. Describing the protocol as
facts and rules also have other benefits i.e. leads to a process-able
knowledge. This knowledge can be transferred as ontology between
systems in KQML format. Since the Prolog language can increase its
knowledge base every time, this method can also be used to learn an
intelligent network.
Abstract: The main objective developed in this paper is to find a
graphic technique for modeling, simulation and diagnosis of the
industrial systems. This importance is much apparent when it is about
a complex system such as the nuclear reactor with pressurized water
of several form with various several non-linearity and time scales. In
this case the analytical approach is heavy and does not give a fast
idea on the evolution of the system. The tool Bond Graph enabled us
to transform the analytical model into graphic model and the
software of simulation SYMBOLS 2000 specific to the Bond Graphs
made it possible to validate and have the results given by the
technical specifications. We introduce the analysis of the problem
involved in the faults localization and identification in the complex
industrial processes. We propose a method of fault detection applied
to the diagnosis and to determine the gravity of a detected fault. We
show the possibilities of application of the new diagnosis approaches
to the complex system control. The industrial systems became
increasingly complex with the faults diagnosis procedures in the
physical systems prove to become very complex as soon as the
systems considered are not elementary any more. Indeed, in front of
this complexity, we chose to make recourse to Fault Detection and
Isolation method (FDI) by the analysis of the problem of its control
and to conceive a reliable system of diagnosis making it possible to
apprehend the complex dynamic systems spatially distributed applied
to the standard pressurized water nuclear reactor.
Abstract: Misalignment and unbalance are the major concerns
in rotating machinery. When the power supply to any rotating system
is cutoff, the system begins to lose the momentum gained during
sustained operation and finally comes to rest. The exact time period
from when the power is cutoff until the rotor comes to rest is called
Coast Down Time. The CDTs for different shaft cutoff speeds were
recorded at various misalignment and unbalance conditions. The
CDT reduction percentages were calculated for each fault and there
is a specific correlation between the CDT reduction percentage and
the severity of the fault. In this paper, radial basis network, a new
generation of artificial neural networks, has been successfully
incorporated for the prediction of CDT for misalignment and
unbalance conditions. Radial basis network has been found to be
successful in the prediction of CDT for mechanical faults in rotating
machinery.
Abstract: In this work we present the modelling of the induction
machine, taking into consideration the stator defects of the induction
machine. It is based on the theory of electromagnetic coupling of
electrical circuits. In fact, for the modelling of stationary defects such
as short circuit between turns in the same phase, we introduce only
in the matrix the coefficients of resistance and inductance of stator
and in the mutual inductance stator-rotor. These coefficients take
account the number of turns in short-circuit deducted from the total
number of turns in the same phase; in this way we obtain the number
of useful turns. In addition, all these faults involved, will be used for
the creation of the database that will be used to develop an automated
system failures of the induction machine.
Abstract: The paper presents an investigation in to the effect of neural network predictive control of UPFC on the transient stability performance of a multimachine power system. The proposed controller consists of a neural network model of the test system. This model is used to predict the future control inputs using the damped Gauss-Newton method which employs ‘backtracking’ as the line search method for step selection. The benchmark 2 area, 4 machine system that mimics the behavior of large power systems is taken as the test system for the study and is subjected to three phase short circuit faults at different locations over a wide range of operating conditions. The simulation results clearly establish the robustness of the proposed controller to the fault location, an increase in the critical clearing time for the circuit breakers, and an improved damping of the power oscillations as compared to the conventional PI controller.
Abstract: As in today's semiconductor industries test costs can make up to 50 percent of the total production costs, an efficient test error detection becomes more and more important. In this paper, we present a new machine learning approach to test error detection that should provide a faster recognition of test system faults as well as an improved test error recall. The key idea is to learn a classifier ensemble, detecting typical test error patterns in wafer test results immediately after finishing these tests. Since test error detection has not yet been discussed in the machine learning community, we define central problem-relevant terms and provide an analysis of important domain properties. Finally, we present comparative studies reflecting the failure detection performance of three individual classifiers and three ensemble methods based upon them. As base classifiers we chose a decision tree learner, a support vector machine and a Bayesian network, while the compared ensemble methods were simple and weighted majority vote as well as stacking. For the evaluation, we used cross validation and a specially designed practical simulation. By implementing our approach in a semiconductor test department for the observation of two products, we proofed its practical applicability.
Abstract: With increasing complexity in electronic systems
there is a need for system level anomaly detection and fault isolation.
Anomaly detection based on vector similarity to a training set is used
in this paper through two approaches, one the preserves the original
information, Mahalanobis Distance (MD), and the other that
compresses the data into its principal components, Projection Pursuit
Analysis. These methods have been used to detect deviations in
system performance from normal operation and for critical parameter
isolation in multivariate environments. The study evaluates the
detection capability of each approach on a set of test data with known
faults against a baseline set of data representative of such “healthy"
systems.
Abstract: As the majority of faults are found in a few of its
modules so there is a need to investigate the modules that are
affected severely as compared to other modules and proper
maintenance need to be done in time especially for the critical
applications. As, Neural networks, which have been already applied
in software engineering applications to build reliability growth
models predict the gross change or reusability metrics. Neural
networks are non-linear sophisticated modeling techniques that are
able to model complex functions. Neural network techniques are
used when exact nature of input and outputs is not known. A key
feature is that they learn the relationship between input and output
through training. In this present work, various Neural Network Based
techniques are explored and comparative analysis is performed for
the prediction of level of need of maintenance by predicting level
severity of faults present in NASA-s public domain defect dataset.
The comparison of different algorithms is made on the basis of Mean
Absolute Error, Root Mean Square Error and Accuracy Values. It is
concluded that Generalized Regression Networks is the best
algorithm for classification of the software components into different
level of severity of impact of the faults. The algorithm can be used to
develop model that can be used for identifying modules that are
heavily affected by the faults.
Abstract: Detection of squirrel cage induction motor (SCIM) broken bars has long been an important but difficult job in the detection area of motor faults. Early detection of this abnormality in the motor would help to avoid costly breakdowns. A new detection method based on particle swarm optimization (PSO) is presented in this paper. Stator current in an induction motor will be measured and characteristic frequency components of faylted rotor will be detected by minimizing a fitness function using pso. Supply frequency and side band frequencies and their amplitudes can be estimated by the proposed method. The proposed method is applied to a faulty motor with one and two broken bars in different loading condition. Experimental results prove that the proposed method is effective and applicable.
Abstract: In this paper the effect of faults in the elements and
parts of discrete event systems is investigated. In the occurrence of
faults, some states of the system must be changed and some of them
must be forbidden. For this goal, different states of these elements are
examined and a model for fail-safe behavior of each state is
introduced. Replacing new models of the target elements in the
preliminary model by a systematic method, leads to a fail-safe
discrete event system.
Abstract: There is lot of work done in prediction of the fault proneness of the software systems. But, it is the severity of the faults that is more important than number of faults existing in the developed system as the major faults matters most for a developer and those major faults needs immediate attention. In this paper, we tried to predict the level of impact of the existing faults in software systems. Neuro-Fuzzy based predictor models is applied NASA-s public domain defect dataset coded in C programming language. As Correlation-based Feature Selection (CFS) evaluates the worth of a subset of attributes by considering the individual predictive ability of each feature along with the degree of redundancy between them. So, CFS is used for the selecting the best metrics that have highly correlated with level of severity of faults. The results are compared with the prediction results of Logistic Models (LMT) that was earlier quoted as the best technique in [17]. The results are recorded in terms of Accuracy, Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The results show that Neuro-fuzzy based model provide a relatively better prediction accuracy as compared to other models and hence, can be used for the modeling of the level of impact of faults in function based systems.
Abstract: In the automotive industry test drives are being conducted
during the development of new vehicle models or as a part of
quality assurance of series-production vehicles. The communication
on the in-vehicle network, data from external sensors, or internal
data from the electronic control units is recorded by automotive
data loggers during the test drives. The recordings are used for fault
analysis. Since the resulting data volume is tremendous, manually
analysing each recording in great detail is not feasible.
This paper proposes to use machine learning to support domainexperts
by preventing them from contemplating irrelevant data and
rather pointing them to the relevant parts in the recordings. The
underlying idea is to learn the normal behaviour from available
recordings, i.e. a training set, and then to autonomously detect
unexpected deviations and report them as anomalies.
The one-class support vector machine “support vector data description”
is utilised to calculate distances of feature vectors. SVDDSUBSEQ
is proposed as a novel approach, allowing to classify subsequences
in multivariate time series data. The approach allows to
detect unexpected faults without modelling effort as is shown with
experimental results on recordings from test drives.
Abstract: The purposes of this research are to study and develop
the algorithm of Thai spoonerism words by semi-automatic computer
programs, that is to say, in part of data input, syllables are already
separated and in part of spoonerism, the developed algorithm is
utilized, which can establish rules and mechanisms in Thai
spoonerism words for bi-syllables by utilizing analysis in elements of
the syllables, namely cluster consonant, vowel, intonation mark and
final consonant. From the study, it is found that bi-syllable Thai
spoonerism has 1 case of spoonerism mechanism, namely
transposition in value of vowel, intonation mark and consonant of
both 2 syllables but keeping consonant value and cluster word (if
any).
From the study, the rules and mechanisms in Thai spoonerism
word were applied to develop as Thai spoonerism word software,
utilizing PHP program. the software was brought to conduct a
performance test on software execution; it is found that the program
performs bi-syllable Thai spoonerism correctly or 99% of all words
used in the test and found faults on the program at 1% as the words
obtained from spoonerism may not be spelling in conformity with
Thai grammar and the answer in Thai spoonerism could be more than
1 answer.
Abstract: Research on damage of gears and gear pairs using
vibration signals remains very attractive, because vibration signals
from a gear pair are complex in nature and not easy to interpret.
Predicting gear pair defects by analyzing changes in vibration signal
of gears pairs in operation is a very reliable method. Therefore, a
suitable vibration signal processing technique is necessary to extract
defect information generally obscured by the noise from dynamic
factors of other gear pairs.This article presents the value of cepstrum
analysis in vehicle gearbox fault diagnosis. Cepstrum represents the
overall power content of a whole family of harmonics and sidebands
when more than one family of sidebands is present at the same time.
The concept for the measurement and analysis involved in using the
technique are briefly outlined. Cepstrum analysis is used for detection
of an artificial pitting defect in a vehicle gearbox loaded with
different speeds and torques. The test stand is equipped with three
dynamometers; the input dynamometer serves asthe internal
combustion engine, the output dynamometers introduce the load on
the flanges of the output joint shafts. The pitting defect is
manufactured on the tooth side of a gear of the fifth speed on the
secondary shaft. Also, a method for fault diagnosis of gear faults is
presented based on order Cepstrum. The procedure is illustrated with
the experimental vibration data of the vehicle gearbox. The results
show the effectiveness of Cepstrum analysis in detection and
diagnosis of the gear condition.
Abstract: Properly sized capacitor banks are connected across induction motors for several reasons including power factor correction, reducing distortions, increasing capacity, etc. Total harmonic distortion (THD) and power factor (PF) are used in such cases to quantify the improvements obtained through connection of the external capacitor banks. On the other hand, one of the methods for assessing the motor internal condition is by the use of Park-s pattern analysis. In spite of taking adequate precautionary measures, the capacitor banks may sometimes malfunction. Such a minor fault in the capacitor bank is often not apparently discernible. This may however, give rise to substantial degradation of power factor correction performance and may also damage the supply profile. The case is more severe with the fact that the Park-s pattern gets distorted due to such external capacitor faults, and can give anomalous results about motor internal fault analyses. The aim of this paper is to present simulation and hardware laboratory test results to have an understanding of the anomalies in harmonic distortion and Park-s pattern analyses in induction motors due to capacitor bank defects.
Abstract: The numerous qualities of squirrel cage induction
machines enhance their use in industry. However, various faults can
occur, such as stator short-circuits and rotor failures.
In this paper, we use a technique based on the spectral analysis of
stator current in order to detect the fault in the machine: broken rotor
bars. Thus, the number effect of the breaks has been highlighted. The
effect is highlighted by considering the machine controlled by the
Direct Torque Control (DTC). The key to fault detection is the
development of a simplified dynamic model of a squirrel cage
induction motor taking account the broken bars fault and the stator
current spectrum analysis (FFT).
Abstract: One major difficulty that faces developers of
concurrent and distributed software is analysis for concurrency based
faults like deadlocks. Petri nets are used extensively in the
verification of correctness of concurrent programs. ECATNets are a
category of algebraic Petri nets based on a sound combination of
algebraic abstract types and high-level Petri nets. ECATNets have
'sound' and 'complete' semantics because of their integration in
rewriting logic and its programming language Maude. Rewriting
logic is considered as one of very powerful logics in terms of
description, verification and programming of concurrent systems We
proposed previously a method for translating Ada-95 tasking
programs to ECATNets formalism (Ada-ECATNet) and we showed
that ECATNets formalism provides a more compact translation for
Ada programs compared to the other approaches based on simple
Petri nets or Colored Petri nets. We showed also previously how the
ECATNet formalism offers to Ada many validation and verification
tools like simulation, Model Checking, accessibility analysis and
static analysis. In this paper, we describe the implementation of our
translation of the Ada programs into ECATNets.
Abstract: Power transformer consists of components which are
under consistent thermal and electrical stresses. The major
component which degrades under these stresses is the paper
insulation of the power transformer. At site, lightning impulses and
cable faults may cause the winding deformation. In addition, the
winding may deform due to impact during transportation. A
deformed winding will excite more stress to its insulating paper thus
will degrade it. Insulation degradation will shorten the life-span of
the transformer. Currently there are two methods of detecting the
winding deformation which are Sweep Frequency Response
Analysis (SFRA) and Low Voltage Impulse Test (LVI). The latter
injects current pulses to the winding and capture the admittance
plot. In this paper, a transformer which experienced overheating and
arcing was identified, and both SFRA and LVI were performed.
Next, the transformer was brought to the factory for untanking. The
untanking results revealed that the LVI is more accurate than the
SFRA method for this case study.
Abstract: Faults in a network may take various forms such as hardware/software errors, vertex/edge faults, etc. Folded hypercube is a well-known variation of the hypercube structure and can be constructed from a hypercube by adding a link to every pair of nodes with complementary addresses. Let FFv (respectively, FFe) be the set of faulty nodes (respectively, faulty links) in an n-dimensional folded hypercube FQn. Hsieh et al. have shown that FQn - FFv - FFe for n ≥ 3 contains a fault-free cycle of length at least 2n -2|FFv|, under the constraints that (1) |FFv| + |FFe| ≤ 2n - 4 and (2) every node in FQn is incident to at least two fault-free links. In this paper, we further consider the constraints |FFv| + |FFe| ≤ 2n - 3. We prove that FQn - FFv - FFe for n ≥ 5 still has a fault-free cycle of length at least 2n - 2|FFv|, under the constraints : (1) |FFv| + |FFe| ≤ 2n - 3, (2) |FFe| ≥ n + 2, and (3) every vertex is still incident with at least two links.